### Replication of "Who Will Defend Democracy? Evaluating Tradeoffs in Candidate Support Among Partisan Donors and Voters"
#### John Carey, Katherine Clayton, Gretchen Helmke, Brendan Nyhan, Mitchell Sanders, Susan Stokes
#### Forthcoming, Journal of Elections, Public Opinion, and Parties.

***

#### Description
Scholars and pundits fear that the American public's commitment to democracy is declining and that citizens are willing to embrace candidates who would trample democratic principles. We examine whether violations of those principles generate resistance from both voters and top campaign donors and whether such resistance extends across partisan lines. In a conjoint survey experiment, we investigate how regular citizens and donor elites trade off partisanship, policy positions, and support for democratic values when choosing between hypothetical political candidates. Our findings indicate that both citizens and donors punish candidates who endorse violations of democratic principles, irrespective of the candidate's party. However, when presented with candidates who support voter identification laws that threaten voter access to the franchise, subjects react very differently, depending on the candidate's party identification. This polarized response is especially strong among donors. These results suggest that the public and donors may sometimes be willing to forgive transgressions against democratic norms that align with their partisan and policy preferences.

#### Files included in this package:

- `_democracy-conjoint.Rproj` -- the directory with this file should be the working directory
- `_README.html` -- generated by `README.md`
- `_README.md` (this file)
- codebooks (folder)
    - `BLW_Wave_8_Conjoint_-_Donors.docx`
    - `MICH0038_B_codebook.pdf`
    - `MICH0033_B_codebook.pdf`
- data (folder)
    - `BLW_Wave_8_Conjoint__Donors.csv`
    - `candidate_names.xlsx`
    - `MICH0033_B_output.sav`
    - `MICH0038_B_output.dta`
- figures (folder)
    - All files are generated by the scripts.
    - _csv (subfolder)
        - All files are generated by the scripts.
- functions (folder)
    - `compare_AMCEs_no_N_labels.R`
    - `compare_AMCEs.R`
    - `estimate_AMCEs.R`
    - `order_attributes_levels.R`
    - `plot_AMCEs_comparison.R`
    - `plot_AMCEs.R`
    - `read_Survey.R`
    - `set_my_ggtheme_legend.R`
    - `set_my_ggtheme.R`
    - `specify_color_palette.R`
    - `specify_parameters.R`
- scripts (folder)
    - `01_read_data_donors.R`
    - `02_read_data_public.R`
    - `03_calculate_AMCEs.R` 
    - `04_linear_regression.R`
    - `05_plot_linearity.R`
    - `06_summary_statistics.R`
    - `07_additional_subgroups.R`
    - `08_party_differences.R`
    - `09_october_read_data.R`
    - `10_october_calculate_AMCEs.R`
    - `11_october_party_differences.R`
    - `12_marginal_means_public.R`
    - `13_marginal_means_donors.R`
- tables (folder) 
    - All files are generated by the scripts.
- temp (folder)
    - All files are generated by the scripts.
    
#### Remarks:

- The `candidate_names.xlsx` spreadsheet was created manually in Excel and includes information from Butler and Homola's (2017) paper in *Political Analysis.* See Table SA.15 in the Supplementary Materials, which are available for download at: <https://www.cambridge.org/core/journals/political-analysis/article/an-empirical-justification-for-the-use-of-racially-distinctive-names-to-signal-race-in-experiments/DBC39F875F2DC0F65E7140FC721CE1EB/share/85b92224c375e025ca057b7367beb5997e1943fe>

#### Program: 

`R` (version 3.6.2)

#### Additional programs required: 

- `tidyverse` (version 1.2.1)
- `readxl` (version 1.2.0)
- `haven` (version 2.1.0.9000)
- `cjoint` (version 2.1.0)
- `ggthemes` (version 4.1.1)
- `estimatr` (version 0.18.0)
- `broom` (verion 0.5.2)
- `directlabels` (version 2018.05.22)
- `scales` (version 1.0.0)
- `stargazer` (version 5.2.2)

#### Process of replication:

##### Main analysis: 

1. Run `01_read_data_donors.R`
2. Run `02_read_data_public.R`
3. Run `03_calculate_AMCEs.R` 
4. Run `04_linear_regression.R`
5. Run `05_plot_linearity.R`

##### Appendix:

1. Run `06_summary_statistics.R`
2. Run `07_additional_subgroups.R`
3. Run `08_party_differences.R`
4. Run `09_october_read_data.R`
5. Run `10_october_calculate_AMCEs.R`
6. Run `11_october_party_differences.R`
7. Run `12_marginal_means_public.R`
7. Run `13_marginal_means_public.R`

#### Most recent date of successful replication

July 19, 2020
